Journal of Operations Management 29 (2011) 318–328
Contents lists available at ScienceDirect
Journal of Operations Management
journal homepage: www.elsevier.com/locate/jom
The boundary spanning capabilities of purchasing agents in buyer–supplier
trust development
Chun Zhang a,∗ , Sridhar Viswanathan b,1 , John W. Henke Jr. c,2
a
School of Business Administration, University of Vermont, Burlington, VT 05405, United States
Eli Broad School of Business, Michigan State University, East Lansing, MI 48824, United States
c
School of Business Administration, Oakland University, Rochester, MI 48309, United States
b
a r t i c l e
i n f o
Article history:
Received 25 November 2009
Received in revised form 8 June 2010
Accepted 1 July 2010
Available online 2 August 2010
Keywords:
Boundary spanners
Supplier relations
Trust
a b s t r a c t
This study examines how individual purchasing agents function as boundary spanners with suppliers
to influence trust development in themselves and the buying firms that employ them. Building upon
boundary theory and supply chain cooperation research, we identify three boundary spanning capabilities
of purchasing agents and empirically test how these capabilities shape buyer–supplier trust development.
Using two samples of data collected from suppliers in the automotive industry and food industry, we
found that a purchasing agent’s effectiveness in strategic communication with suppliers affects a supplier’s
trust in the buying firm, while an agent’s professional knowledge and ability to reach compromises with
suppliers affect a supplier’s trust in the purchasing agent representing the firm. Trust in the purchasing
agent in turn affects trust in the buying firm. Theoretical and managerial implications are discussed.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Collaborative buyer–supplier relations are a major source
of competitive advantage for businesses operating in industrial
markets (e.g. Takeishi, 2001; Paulraj et al., 2008). Members of
collaborative buyer–supplier relations share strategic directions,
which helps define the roles and responsibilities of supply chain
members (Ireland and Webb, 2007). Additionally, members of
collaborative relations share high levels of trust (MacDuffie and
Helper, 2006). In fact, trust is considered the single most important variable influencing interpersonal and inter-organizational
behavior (Kiessling et al., 2004). For a variety of reasons, trusting buyer–supplier relationships lead to reduced transaction costs
and improved supply chain efficiencies (e.g. Zaheer et al., 1998;
MacDuffie and Helper, 2006; Ireland and Webb, 2007).
Despite the sustainable competitive advantages generated by
trusting buyer–supplier relations, developing trust is difficult. The
personnel chosen to build trust between supply chain members are
of great importance (MacDuffie and Helper, 2006). This is understandable, because the establishment and maintenance of trusting
working relations rely on the individuals who regularly interact
with one another across organizational boundaries (Perrone et al.,
∗ Corresponding author. Tel.: +1 802 656 4116; fax: +1 802 656 8279.
E-mail addresses: [email protected] (C. Zhang), [email protected]
(S. Viswanathan), [email protected] (J.W. Henke Jr.).
1
Tel.: +1 281 804 7677.
2
Tel.: +1 248 644 7690; fax: +1 248 644 7488.
0272-6963/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jom.2010.07.001
2003; Ireland and Webb, 2007). These boundary spanners include
purchasing agents who develop relationships with individuals of
other firms, specifically suppliers. These relations at the individual
level provide “a portal for broader communications between organizations” that generate familiarity and trust (Kiessling et al., 2004,
p. 99).
As points of contact with the outside world, boundary spanners
influence trust development in individuals as well as in organizations (Doney and Cannon, 1997; Zaheer et al., 1998; Perrone
et al., 2003). Perrone et al. (2003) suggest that not only do job
functions affect boundary spanner behaviors, but boundary spanners also proactively shape their job functions. The formal job
functions specified by their organization can constrain boundary
spanner behavior, suggesting that organizations are at least partly
responsible for boundary spanner job performance. However, the
job functions of boundary spanners are also subject to interpretation and are shaped by the individuals performing the functions
(Perrone et al., 2003). It is thus reasonable to believe that boundary spanners influence the trust outside organizations place in
the firms they represent, as well as the trust the outside organizations place in the boundary spanners themselves. Thus, it is
important to understand how the capabilities individual boundary
spanners demonstrate in performing their job functions affect the
trust boundary spanners generate both in themselves and in the
firms they represent.
This research area has been of interest to a number of strategy
and operations management scholars (Stanley and Wisner, 2001;
Kiessling et al., 2004; MacDuffie and Helper, 2006; Ireland and
Webb, 2007). Ireland and Webb (2007) considered the role bound-
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
ary spanners play in influencing trust and power in strategic supply
chains. Other operations management researchers have also provided qualitative insights into the role boundary spanners play in
building trusting relationships with suppliers (Stanley and Wisner,
2001; Kiessling et al., 2004; MacDuffie and Helper, 2006). Nevertheless, limited research has identified the boundary spanning
capabilities that influence trust development. No empirical tests
examine how boundary spanning capabilities influence trust development at both the interpersonal and inter-organizational levels
(Doney and Cannon, 1997; Zaheer et al., 1998; Perrone et al., 2003).
Such tests are important in understanding how trust is developed
and maintained (MacDuffie and Helper, 2006).
To address this gap in the literature, we investigate how the
boundary spanning capabilities of purchasing agents influence a
supplier’s trust of purchasing agents and of the buying firms the
agents represent. We choose to study purchasing agents as their
roles have changed significantly over the past two decades from
“transactions-oriented order processors to supply managers with
an emphasis on supply chain management” (Stanley and Wisner,
2001). Because purchasing agents can significantly influence a buying firm’s reputation and image (Stanley and Wisner, 2001), buying
firms are increasingly relying on them to build cooperative relations
with major suppliers and to encourage supplier-developed innovations (MacDuffie and Helper, 2006; Zhang et al., 2009). Specifically,
purchasing agents carry out a broad range of activities from representing their firms’ strategic goals and intent, to using their
expertise to facilitate buyer–supplier collaborations (Aldrich and
Herker, 1977; Perrone et al., 2003). Building on boundary theory
(e.g., Adams, 1976; Aldrich and Herker, 1977) and existing research
on supply chain cooperation, we focus on the purchasing agent’s
capabilities most relevant to trust development with suppliers. Furthermore, to increase the external validity of our study findings,
we collect two samples to test our conceptual model: one from the
automotive industry and one from the food industry. We found consistent evidence for the effects of boundary spanning capabilities on
trust development across the two industries. The theoretical basis
for the study and the study findings are presented in the following
sections. We conclude by discussing the theoretical and managerial
implications of the study and directions for future research.
2. Theoretical development
2.1. Trust
Trust has been investigated by a number of researchers in strategy and organizational research (e.g. Bradach and Eccles, 1989; Ring
and Van De Ven, 1992; Zaheer et al., 1998; Perrone et al., 2003)
and supply chain management (e.g. Anderson and Narus, 1990;
Sako and Helper, 1998; Doney and Cannon, 1997; McCutcheon
and Stuart, 2000; Johnston et al., 2004; Gattiker et al., 2007). Trust
is a critical factor in developing cooperative buyer–supplier relationships (e.g. MacDuffie and Helper, 2006; Zaheer et al., 1998).
For example, trust reduces opportunism in downstream supply
chains (Cavusgil et al., 2004), improves supply chain responsiveness (Handfield and Bechtel, 2002), and increases the potential for
beneficial supply chain alliances (McCutcheon and Stuart, 2000).
While no consensus has been reached on a universally accepted
definition of trust, most trust scholars agree that confidence and
belief in the other party’s credibility and goodwill are essential
to conceptualize trust (e.g. Doney and Cannon, 1997; Johnston et
al., 2004; Gattiker et al., 2007). Furthermore, trust scholars (e.g.
McCutcheon and Stuart, 2000; Johnston et al., 2004; Gattiker et
al., 2007; Hill et al., 2009) have hypothesized and established that
the credibility and goodwill aspects of trust are affected in the
same way by the same determinants. In particular, when multiple
319
levels of trust (e.g. interpersonal and inter-organizational levels)
are examined, scholars often combine the credibility and goodwill
dimensions of trust to simplify the conceptual model (Doney and
Cannon, 1997; Zaheer et al., 1998; Perrone et al., 2003). As explained
by Doney and Cannon (1997), “although credibility and benevolence could be conceptually distinct, in business relationships such
as those studied here, they may be so intertwined that in practice
they are operationally inseparable” (p. 43). Therefore, we consider
trust as a unidimensional construct; trust in our study is defined
as a supplier’s confidence and belief in the credibility and goodwill
of an object of trust. The object may be a purchasing agent or the
buying organization represented by the agent.
We investigate both interpersonal trust and interorganizational trust in buyer–supplier relations. While
operations management scholars acknowledge the importance
of interpersonal trust, past research has primarily focused on
inter-organizational trust (e.g. Johnston et al., 2004; Gattiker et al.,
2007; Hill et al., 2009). Zaheer et al. (1998) have suggested though
that “although inter-organizational trust may appear to be the
more important influence in relational exchange, interpersonal
trust must also be considered for its effects on inter-organizational
trust. . . . Simply aggregating interpersonal trust as a proxy for
inter-organizational trust ignores the influence of social context in
the form of individuals’ interactions and organizational rules that
constrain and orient its members” (p. 154). To understand how the
job performance of purchasing agents shapes trust in individuals
and firms, we investigate a supplier’s trust in a purchasing agent
as well as in the buying firm the agent represents.
As one type of boundary spanner, purchasing agents play a
critical role in building supplier trust by reducing the risks suppliers perceive when working with a powerful buyer (Perrone et al.,
2003; Ireland and Webb, 2007). As Perrone et al. (2003) suggest,
trust is most relevant when risk and uncertainties are involved in
buyer–supplier relations. In an industrial supply market, a buying
firm often deals with numerous suppliers depending on the complexity of the goods being acquired (Takeishi, 2001; MacDuffie and
Helper, 2006). This competitive situation creates risk and uncertainty for suppliers who depend on the buying firms and their
purchasing agents for continued business. The risk for suppliers
begins with the policies and strategies set by the buying firm that
can affect the supplier firm’s costs and profitability (IBM Business
Consulting Services, 2004). The purchasing agent, as the primary
commercial contact with the supplying firm, mitigates this risk
through information provided to supplier salespeople, which in
turn affects the supplier’s trust in the purchasing agent and the
buying firm (Ireland and Webb, 2007). Trust in purchasing agents
and buying firms is critical because suppliers who do not trust their
customers are unlikely to make long-term investments to support
future business with the buyer (Doney and Cannon, 1997; Perrone
et al., 2003).
2.2. Boundary spanning capabilities of purchasing agents
Boundary theory argues that a central task of organizations is
to manage their boundaries with other organizations that supply critical resource inputs or are responsible for the disposal of
their outputs (e.g., Aldrich and Herker, 1977; Stock, 2006). To fulfill
this task, organizations assign individual employees to perform a
varieties of boundary spanning functions such as processing information; maintaining the image of the organization; using expertise
to influence external entities; and representing the perceptions,
expectations, and ideas of each side to the other (e.g. Aldrich and
Herker, 1977; Friedman and Podolny, 1992; Ireland and Webb,
2007). Among these boundary spanning functions, those related
to representing the buying firm to external suppliers are most relevant to building trust in supply chain relationships (Perrone et
320
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
Fig. 1. A model of boundary spanning capabilities and trust development.
al., 2003; Stock, 2006). We classify these external representation
functions broadly into three categories. The first is that boundary
spanners may communicate information from the core of the organization to the environment (Aldrich and Herker, 1977; Perrone
et al., 2003). Second, boundary spanners may use their knowledge
and expertise to influence the external environment (Aldrich and
Herker, 1977; Doney and Cannon, 1997; Perrone et al., 2003). Third,
boundary spanners can strike a compromise between their organization and the environment in which it is operating (Aldrich and
Herker, 1977; Friedman and Podolny, 1992).
In performing these boundary spanning functions, individual
employees demonstrate boundary spanning capabilities that allow
them to effectively manage their firm’s interactions with other entities and facilitate collaborations across organizational boundaries
(Stock, 2006; MacDuffie and Helper, 2006). For the purpose of this
study, we define boundary spanning capabilities as the job performance of individual purchasing agents carrying out boundary
spanning functions. Based on the three broad types of boundary
spanning functions discussed earlier, we conceptualize boundary
spanning capabilities of purchasing agents as composed of three
major types: (1) strategic communication capability, or the effectiveness of purchasing agents at presenting their firm’s strategic
intention and goals to suppliers (Ireland and Webb, 2007; Perrone
et al., 2003); (2) professional knowledge, the knowledge possessed
by purchasing agents that enables them to competently carry out
their job responsibilities (Doney and Cannon, 1997); and (3) ability
to reach compromise with suppliers, or the effectiveness of purchasing agents at mediating and overcoming differences that arise
between their firms and the suppliers they interact with (Friedman
and Podolny, 1992).
While many different types of boundary spanning capabilities can be examined, we chose to focus on the capabilities most
conducive to facilitating buyer–supplier cooperation and trustbuilding. The boundary spanning capabilities of purchasing agents
are critical to establishing and maintaining supply chain relationships (e.g., Aldrich and Herker, 1977; Perrone et al., 2003; Ireland
and Webb, 2007). The three boundary spanning capabilities of purchasing agents capture not only the effectiveness of purchasing
agents at representing the policies and strategic directions delineated by the firm, but also their abilities to proactively shape the
interaction environment between buyers and suppliers (Aldrich
and Herker, 1977; Perrone et al., 2003).
We acknowledge that boundary spanning capabilities of purchasing agents can be influenced by the intrinsic dispositional traits
of the individual agent (Smith et al., 2009). The focus of our study,
however, is not on a purchasing agent’s dispositional traits as these
are difficult to measure (Smith et al., 2009) and, more importantly,
may not be directly related to trust-building at multiple levels.
Instead, we limit our focus to the capabilities of purchasing agents
as expressed by their performance of boundary spanning functions.
These capabilities can be considered as expressed dispositional
traits of purchasing agents in performing boundary spanning functions, and are specific to a relationship between a purchasing agent
and a salesperson. As a purchasing agent may have different experiences with various supplier salespersons, the assessment of a
purchasing agent’s ability to perform boundary spanning functions
can vary among supplier relationships; an agent may inspire the
trust of one salesperson, but not the trust of others.
3. Conceptual model and hypotheses
We propose that a purchasing agent’s effectiveness in strategic
communication enhances trust in the buying firm, while his or her
professional knowledge and ability to reach compromise enhances a
supplier’s trust in the purchasing agent, which is then transferred
to trust in the buying firm. Although all three types of boundary
spanning capabilities of purchasing agents could be attributed to a
buying firm’s training and corporate policies and therefore trigger
trust-building processes in both individuals and firms, this research
focuses on the primary effects of a certain boundary spanning capability (Doney and Cannon, 1997). The proposed model is presented
in Fig. 1.
3.1. Strategic communication
As boundary spanners, purchasing agents act as representatives
of their firm’s strategic intention and goals (Ireland and Webb,
2007; Perrone et al., 2003). Ireland and Webb (2007) propose that
“while just-in-time and other electronic data interchange systems
link partnering firms technologically and operationally, boundary
spanning individuals integrate firms strategically” (p. 493). When
a purchasing agent communicates to a supplier their firm’s longterm vision, it enables a supplier to determine how the future
plans of the buying firm are likely to affect it. Such information
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
helps the supplier determine its internal investment priorities in
terms of innovation, capacity, product type, and location decisions (Takeishi, 2001; Paulraj et al., 2008; Zhang et al., 2009).
Furthermore, through communication of strategic information, a
purchasing agent enables a supplier to determine how it can help
influence the buying firm’s future purchase decisions. For example, a purchasing agent can communicate to a supplier which
innovations and new technologies the buying firm needs. These
communications involve information that is of a long-term, strategic nature and can be performed for each product type supplied by
a supplier to the buying firm (Kraljic, 1983).
Strategic communications increase a supplier’s trust in the buying firm by reducing the risks and uncertainties surrounding a
supplier’s long-term plans for servicing the buying firm (Zhang et
al., 2009). Through strategic communication of purchasing agents,
the strategic goals and intentions of buying firms are revealed,
which helps to establish shared norms and values between buyers and suppliers (Ireland and Webb, 2007; Paulraj et al., 2008). As
Ireland and Webb (2007) note, “Given that strategic supply chains
are composed of firms with idiosyncratic objectives and contexts,
these [strategic communication] roles offer transparency to individual firm actions.” (p. 493).
The trust generated by strategic communication is directed
toward the buying firm responsible for approving the communication rather than the purchasing agent for a variety of reasons. The
buying firm the purchasing agent represents delineates the content of strategic communication (Perrone et al., 2003). As such, a
purchasing agent primarily accepts their role of communicating
strategy passively as opposed to actively choosing the content of
their strategic communication (Perrone et al., 2003). This capability
directly links a purchasing agent to the buying firm in the supplier’s mind as he or she is essentially functioning as the conveyor of
the buying firm’s message without needing to add significant personal interpretations to the message (e.g. Aldrich and Herker, 1977;
Ireland and Webb, 2007). Because purchasing agents are unlikely
to exercise a significant amount of discretion when communicating the strategic objectives and intent of the buying organization
they represent, a supplier firm is likely to assign responsibility for
the communications delivered to the buying firm and thus manifest trust toward the buying firm rather than the purchasing agent
delivering the message (Perrone et al., 2003). Perrone et al. (2003)
observe that trust in the buying firm increases significantly as a
purchasing agent carries out roles strictly delineated by the buying
firm. Accordingly, we hypothesize that:
H1. Purchasing agents’ strategic communication is positively
related to a supplier’s trust in a buying firm.
321
agent possessing it more powerful (Perrone et al., 2003). In other
words, a purchasing agent’s professional knowledge reveals a purchasing agent’s competence as an individual independent of the
buying firm (Doney and Cannon, 1997). It allows the supplier firm
to assess the power and knowledge of a purchasing agent to manage people in the buying organization to carry out the agreements
he or she has reached with the supplier representative (Perrone et
al., 2003).
A supplier is more likely to develop trust in the purchasing
agent when they perceive the purchasing agent as being competent and able to deliver on his or her promises (Doney and Cannon,
1997; Perrone et al., 2003). This observation is consistent with
the finding that a boundary spanner’s expertise results in a positive evaluation of the boundary spanner by his or her counterparts
(Biong and Selnes, 1996). Furthermore, when a purchasing agent is
knowledgeable about a supplier’s product and business, this forms
a common knowledge base the two firms can share (Schroeder et
al., 2008). Suppliers are more likely to feel understood and valued
when they perceive a purchasing agent as knowledgeable about
their product(s) and business. In this sense, a purchasing agent’s
professional knowledge serves as a positive indication to the supplier that the purchasing agent is inclined to facilitate joint-goal
achievement between the two parties and is thus trustworthy
(Johnson and Johnson, 1972; Doney and Cannon, 1997).
Boundary spanner professional knowledge resides within the
individual purchasing agent rather than the buying firm. In contrast
to strategic communication, where purchasing agents passively
accept the role directed by the firm they represent, professional
knowledge enables purchasing agents to proactively shape the
boundary spanning role they perform (Spekman, 1979; Doney
and Cannon, 1997). Spekman (1979) reported that knowledge and
experience are major reasons a boundary spanner is considered
influential in decision-making. As such, professional knowledge
gives purchasing agents higher professional status and allows them
to exercise their expert power within the buying firm (Spekman,
1979; Tushman and Scanlan, 1981). Suppliers perceive this expert
power as an indication that a purchasing agent has decision-making
autonomy within their organization, which makes the supplier
more likely to trust the purchasing agent (Perrone et al., 2003).
Doney and Cannon (1997) found that a salesperson’s expertise significantly increases trust in the salesperson. Similarly, Perrone et
al. (2003) found that a purchasing agent’s tenure within an organization significantly increases a supplier’s trust in the purchasing
agent, as greater tenure is associated with greater knowledge. Thus,
we hypothesize:
H2. Purchasing agents’ professional knowledge is positively
related to a supplier’s trust in a purchasing agent.
3.2. Professional knowledge
3.3. Ability to compromise
Purchasing agents also shape relations with suppliers through
their professional knowledge, i.e., the knowledge that enables them
to competently carry out their responsibilities. Purchasing agents
can use their professional knowledge to shape interactions with
suppliers and thus gain greater control over the environment external to the firm (Doney and Cannon, 1997; Perrone et al., 2003). An
organization may have a great deal of knowledge resources at the
organizational level. Such knowledge may not always be accessible to the individual purchasing agent, who may need a wide
breadth of knowledge to make intelligent decisions (Aldrich and
Herker, 1977). A purchasing agent can gain power within the buying organization by accumulating knowledge about the products,
processes, and working style of the buying and supplying organizations (Perrone et al., 2003). Over time, the professional knowledge
accumulated by a purchasing agent may become a scarce resource,
which makes the knowledge more valuable and the purchasing
A purchasing agent’s ability to compromise deals with the
mediation activities purchasing agents undertake to overcome differences that arise between their firm and the suppliers from whom
they are acquiring goods and services. Perrone et al. (2003) propose that a purchasing agent’s ability to resolve potential conflicts
between buying and supplying organizations and reach mutually
acceptable solutions demonstrates the purchasing agent’s intention and ability to consider the interests of both collaborative
parties. It also represents a discretionary behavior on the part of the
purchasing agent to maintain the spirit of commitment (Perrone
et al., 2003). Indeed, these activities can be a challenge for the
purchasing agent, who as a representative of the buying firm is
often assumed by the salesperson to be acting solely in the buying
firm’s best interest. Successful mediation under these conditions,
which can easily degenerate into contentious adversarial meetings,
322
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
requires that the purchasing agent demonstrate an understanding
of and empathy for the concerns of the supplier, while striving to
reach equitable solutions for both the buying and supplying firms
(Friedman and Podolny, 1992; MacDuffie and Helper, 2006). Furthermore, purchasing agents must also act as advocates for the
supplier within their firm to help ensure that decisions by other
employees of the buying firm, who are looking out for the best interests of themselves and their firm, do not unduly negatively impact
the supplying firm (Friedman and Podolny, 1992). By balancing the
interests of the supplier with the interests of their firm, purchasing
agents can effectively mediate between the buying and supplier
firms, reducing antagonism and adversarial feelings—particularly
on the part of the supplier. The purchasing agent’s ability to compromise in mediation situations deals with engaging an external
organization, i.e., the supplier, in such a way as to build the perception that the purchasing agent cares about the needs of the supplier
and thus can be trusted.
As with professional knowledge, a supplier’s trust in the purchasing agent rather than trust in the buying firm increases when
a purchasing agent consistently demonstrates the ability to reach
compromise with suppliers. Although a purchasing agent’s ability
to compromise can be influenced by the firm for which the individual works, it is rationally attributed to individuals (Friedman
and Podolny, 1992). Friedman and Podolny (1992) explains that
for individual boundary spanners to be trusted by both the firm
they represent and the parties with whom they interact, they
need to substantively engage in compromise behaviors. In a similar
line of thought, MacDuffie and Helper (2006) propose that a purchasing agent’s ability to reach a compromise is highly critical to
managing supplier relations in the automotive industry. MacDuffie
and Helper (2006) further observe that a key determinant of the
long-term competitiveness of buyers in the automotive industry is
the ability of purchasing agents to reach mutually beneficial costreduction solutions with suppliers. Doing so requires purchasing
agents to exercise their discretionary decision-making autonomy
when managing relationships with suppliers (Perrone et al., 2003).
As such, a supplier is likely to place their trust in an individual
purchasing agent rather than the buying firm with whom the purchasing agent is associated. Thus, we hypothesize:
H3. Purchasing agents’ ability to reach compromise is positively
related to a supplier’s trust in a purchasing agent.
3.4. Trust in purchasing agents and trust in buying firms
A supplier’s trust in a purchasing agent and in the buying firm
employing the agent are likely to be intertwined as one reinforces
the other (Doney and Cannon, 1997; Zaheer et al., 1998). As the
primary buying firm commercial contact, the purchasing agent represents an important source of trust for the supplier (MacDuffie and
Helper, 2006; Doney and Cannon, 1997). The repeated interactions
between purchasing agents and supplier salespeople enable the
supplying firm to develop trust in the purchasing agent (Jap and
Anderson, 2003), which in turn strengthens the overall relationships between the buying and selling firms (Bradach and Eccles,
1989; Doney and Cannon, 1997; Zaheer et al., 1998). As such, trust
in the purchasing agent can serve as an antecedent of trust in the
buying firm the purchasing agent represents.
The influence between trust in the purchasing agent and trust in
the buying firm may occur in the other direction as well (Doney and
Cannon, 1997; Zaheer et al., 1998). For example, trust can transfer
from a well-know organization to a less well-know group or individuals who represent the organization (Doney and Cannon, 1997).
Also, a salesperson’s previous interactions with various members of
the buying firm can be used to infer to what extent a new unknown
purchasing agent can be trusted. Using data collected from 200
purchasing managers, Doney and Cannon (1997) found a reciprocal causation between trust in salespeople and trust in supplier
firms. Though we examine purchasing agents and buying firms, this
still supports the notion of trust in a firm influencing trust in the
boundary spanners they employ. As such, we hypothesize:
H4a. Supplier firm trust in the purchasing agent increases supplier
firm trust in the buying firm.
H4b. Supplier firm trust in the buying firm increases supplier firm
trust in the purchasing agent.
4. Method
4.1. Research setting and data collection
The data were collected from production goods suppliers of
two major global manufacturing firms in the automotive and food
industries. We chose the automotive and food industries because
they differ significantly in the nature of products procured and the
degree of innovation required in the upstream supply chain. As a
result, diverse practices exist in managing supplier relationships in
these two industries, which helps to increase the generalizability of
the study findings. The North American-based automotive assembler and European-based processed food manufacturer provided
contact information for their production suppliers. The suppliers
for each firm received minimum annual purchases of $250K from
the automotive assembler or $160K from the food processor. All of
these suppliers were in good standing with the respective buying
firms. We focused on the buyer–supplier relations at the buying situation level. Each buying situation is defined as production goods
in a specific commodity area provided to a specific manufacturing
site by a supplier. For example, for the automotive assembler, a
buying situation can be exhaust systems provided to a GM plant by
a Delphi division. The buying situation was adopted as the unit of
analysis since buyer–supplier relationship management practices
in manufacturing industries vary significantly across buying situations (Laseter and Ramdas, 2002; Takeishi, 2001; Dyer and Hatch,
2006).
Supplier salespeople were used as the key informants for the
study. Our interviews with supplier engineering and sales personnel at various levels indicated that the salespeople in their role of
having commercial responsibility for a specific buying account, are
directly responsible for resource allocation to the account. Furthermore, supplier salespeople are intimately aware of the relationship
management practices the buying firm has implemented with their
firm. In completing the survey questionnaire the supplier salespeople first selected the commodity area(s) for which they have
responsibility and then answered the questions as they related to
the working relations their firm has with the buying firm for the
commodity area(s) for which they have commercial responsibility.
The approach of key informant selection adopted in this study is
consistent with that recommended by Kumar et al. (1993).
A large-scale Internet-based survey of the supplier salespeople
was implemented independently for each buying organization. A
total of six weekly reminders were emailed to each non-responding
salesperson before data collection was completed. We received
contact information for 692 supplier salespeople working for 332
automotive suppliers and contact information for 283 supplier
salespeople working for 153 food suppliers. We asked the supplier
respondent to answer the questions in each questionnaire as they
relate to a single buying situation they have commercial responsibilities for. They were instructed to rate the purchasing agent with
whom they have had the most contact over the past year. For the
automotive industry sample, the purchasing agents are referred
to as “commodity buyers”, and for the food industry sample, the
purchasing agents are referred to as “buyers”. Out of the supplier
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
salespeople contacted (692 in the automotive industry and 332
in the food processing industry), 230 salespeople in the automotive industry completed usable questionnaires corresponding to a
response rate of 33%, and 125 salespeople in the food industry filled
out questionnaires corresponding to a response rate of 44%. These
responses provide completed information on all focal constructs for
345 buying situations in the automotive industry and 143 buying
situations in the food industry. The majority of the responses (66.1%
in the automotive industry; 88.2% in the food processing industry) are provided by an individual supplier salesperson or team
reporting on an individual buying situation to a buying organization. There are instances where an individual supplier salesperson
or team has major responsibility for multiple buying situations (the
remaining 33.9% for the automotive industry and 12.8% for the food
processing industry). In these cases, the individual salesperson or
team provided multiple responses concerning the multiple buying
situations. In all cases, each buying situation is recorded on a single
questionnaire.
We searched for evidence of non-response bias by comparing
early and late responses in terms of all of the constructs included in
the study (Armstrong and Overton, 1977). We found no significant
differences between early and late response groups on these key
variables. The supplier firms sample for the automotive assembler
had been supplying the global manufacturer from 1 to 100 years
with a median of 20 years. The supplier firms sample in the food
industry has been supplying the food processor from 1 to 80 years
with a median of 12 years.
To minimize the potential common method bias introduced by
the use of self-reported data from a single source of information,
we encouraged team consensus when responding to the survey.
Responses provided by team consensus are likely to reduce the bias
associated with any particular individual’s attitude and cognition
on the subject matter. This is in particular important in situations
where an individual respondent is uncertain of some of the questions in the survey. We encouraged the salesperson to whom the
survey questionnaire was directed to involve other supplier personnel who worked with the assembler to assist them in answering
the questions, if they were uncertain of the appropriate response.
By including an item in the survey asking how the survey was completed, we found that 55.5% of the completed questionnaires for
the automotive assembler and 50.2% of the completed questionnaires for the food processor were completed by consensus reached
among multiple supplier personnel who were familiar with their
firm’s working relations with the manufacturer. Of the remaining
completed questionnaires that were answered by a single salesperson, the salespeople tended to work for local suppliers or were
responsible for a single product and were understandably confident of their knowledge of the questionnaire items. We further
examine whether data collected from the two types of responses
produce consistent findings. We split the samples into two groups
by whether the responses were formed by team consensus. We ran
the hypothesized model and found that the results for both groups
are consistent, and there are no significant differences for the coefficient estimates between the two groups. Furthermore, we also
employed Harman’s single factor test to examine whether common
method variance presented a significant threat to the data analysis. We found that the hypothesized measurement models fit the
data significantly better than a single factor model, suggesting that
common method bias is not a significant concern in the current
study.
4.2. Questionnaire instrument
The survey instrument was developed in three stages. First,
a literature review identified a list of potential measures we
could adopt for the study. Second, we conducted interviews with
323
purchasing directors, managers, and purchasing agents in each
manufacturer (36 in the automotive assembler and 25 in the food
processor) and with supplier salespeople who had commercial
responsibility for each manufacturer (13 automotive suppliers and
18 food processor suppliers) to determine the content validity of
the potential measures. These interviews resulted in modifications
of some measures and also identified several new measures. We
revised the survey instrument based on a pre-test of the questionnaire involving additional salespeople who did not help in the
development of the measures (six automotive supplier salespeople
and five food processor supplier salespeople).
4.3. Measures
All measurements were made using a one to five scale (1 = to a
very little extent, 5 = to a very great extent). A supplier’s trust of purchasing agent (BTRUST) is measured by a four-item scale modified
from Perrone et al. (2003). It measures to what degree a supplier
considers a purchasing agent to be (1) overall trustworthy, (2)
has high integrity, (3) respectful of the salesperson and others in
the supplier firm, and (4) reliable in making negotiation-related
promises with the supplier firm. The Cronbach alpha for the fouritem scale is 0.90 for the automotive manufacturer and 0.87 for the
food processor.
A supplier’s trust of buying firm (FTRUST) is measured by a fouritem scale modified from Perrone et al. (2003). It captures the
degree to which a supplier considers the buying firm (1) to be overall trustworthy, (2) to be fair in dealing with the supplier (3) to live
up to the spirit of its commitments, and (4) to treat the supplier as
a valued partner. The Cronbach alpha for the construct is 0.91 for
the automotive manufacturer and 0.87 for the food processor.
The scales for the three types of boundary spanning capabilities, strategic communication, professional knowledge, and ability
to reach compromise were developed for this study based on
ideas presented in Aldrich and Herker (1977) and Perrone et al.
(2003) and field interviews. Purchasing agent strategic communication (BCOM) was measured by a 4-item scale that captures the
degree to which a purchasing agent is effective at explaining (1)
the supplier’s role in the buying firm’s long-term overall purchasing
strategy, (2) the buying firm’s purchasing strategy for the products
supplied by the supplier, and (3) the importance of the supplier
firm in bringing innovation and/or new technologies to the buying firm. We learned through field interviews that the contents of
strategic communication can be industry-specific, so we included
a 4th item for the automotive industry, which measures the degree
to which a purchasing agent is effective at explaining the buying
firm’s supplier suggestion program. For the food industry, the 4th
item measures the degree to which a purchasing agent is effective
at explaining contracts and agreements to the supplier. The Cronbach alpha for the construct is 0.89 and 0.88 for the automotive and
food manufacturers, respectively.
Purchasing agent professional knowledge (BKNOW) measures
the degree to which a purchasing agent has sufficient (1) commercial knowledge, (2) general product technical knowledge, (3)
engineering or manufacturing process knowledge, (4) knowledge
of the supplier’s product, and (5) knowledge of the supplier’s capabilities. The Cronbach alpha for the five-item construct is 0.92 and
0.90 for the automotive and food manufacturers, respectively.
Purchasing agent ability to reach compromise (BCOMPS) measures the degree to which a purchasing agent (1) reaches solutions
that are equitable to both the buying firm and the supplier firm
when differences arise, (2) generally considers the supplier’s commercial and financial interests, and (3) acts as the supplier’s
advocate with the buying firm’s engineering division. The Cronbach alpha for the three-item construct is 0.84 for the automotive
manufacturer and 0.75 for the food manufacturer.
324
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
Table 1
Reliability construct measurements and validity assessment.
Construct measures
Standard loading
(automotive industry)
Standard loading (food
industry)
Purchasing agent strategic communication capability
The purchasing agent is effective at:
1. Explaining the buying firm’s purchasing strategy for the products supplied by your firm·
2. Explaining your firm’s role in the buying firm’s long-term overall purchasing strategy
3. Explaining the importance of your firm bring innovation and new technologies to the
buying firm
4. Explaining the buying firm’s supplier suggestion program/contracts or agreements to
supply
Purchasing agent professional knowledge
1. Commercial knowledge
2. General product technical knowledge
3. Engineering/manufacturing process knowledge
4. Knowledge of your firm’s products·
5. Knowledge of your firm’s capabilities·
˛ = .89
˛ = .88
.863
.890
.830
.874
.883
.777
.628
.792
˛ = .92
.791
.871
.874
.838
.801
˛ = .90
.784
.847
.677
.862
.875
Purchasing agent ability to reach compromise
1. The purchasing agent generally strives to reach solutions that are equitable to both the
buying firm and your firm when differences arise
2. The purchasing agent acts as your firm’s advocate with the buying firm’s engineering
division
3. The purchasing agent generally considers your firm’s commercial and financial interests
˛ = .84
.805
˛ = .75
.820
.721
.508
.893
.631
A supplier firm’s trust of a purchasing agent
1. The purchasing agent is respectful of you and others in your firm
2. The purchasing agent has high integrity
3. The purchasing agent is overall trustworthy
4. The purchasing agent is reliable in making negotiation-related promises with our firm·
˛ = .91
.811
.945
.939
.646
˛ = .87
.840
.910
.943
.576
A supplier firm’s trust of a buying firm
1. The buying firm treat your firm as a valued supplier
2. The buying firm lives up to the spirit of its commitments
2. The buying firm is fair in its dealing with your firm
3. The buying firm is overall trustworthy·
˛ = .91
.809
.829
.888
.893
˛ = .87
.791
.771
.862
.816
We include two control variables that are relevant to
buyer–supplier relationship management: supplier sales dependency and relationship length. Supplier sales dependency is
measured as the approximate share of a supplier firm’s annual sales
to the buying company; relationship length is measured by the
number of years a supplier firm has been supplying similar products
to the buying company.
5. Data analysis and results
5.1. Measurement model evaluation
We evaluate the measurement models for the two samples
using the following methods. First, we evaluate reliability and the
construct validity of independent and dependent constructs using
Cronbach’s alpha coefficient and confirmatory factor analysis (CFA)
(EQS 6.1). The Cronbach’s alpha coefficients for all the constructs
for both samples exceed the 0.7 cutoff value, therefore construct
reliability is established. The pre-specified CFA model provides an acceptable fit (2 (160) = 701.274, CFI = 0.907, IFI = 0.907,
GFI = 0.827, RMSEA = 0.099 for the automotive industry; (2
(160) = 356.522, CFI = 0.908, IFI = 0.909, GFI = 0.805, RMSEA = 0.093).
All factor loadings were statistically significant at the 5% level and
exceed the arbitrary 0.5 standard. We tested for discriminant validity by comparing an unconstrained model with a model with the
correlation between each pair of constructs constrained to equal
one (Bagozzi et al., 1991; Zaheer et al., 1998). We found that in all
cases the cross-construct correlations were significantly different
from 1.0, which suggests the presence of discriminant validity. In
general, these results provide support for construct validity for the
measures employed in the study. The results for construct validity
and reliability are reported in Table 1.
5.2. Structural model evaluation
After reliability and construct validity were established for the
measurement models, structural path parameters were estimated.
We began by examining the reciprocal relations between trust
in the buying firm and trust in the purchasing agent. After the
direction of the relationship between the two levels of trust was
established, we proceeded to test the hypothesized relationships
between the boundary spanning capabilities and trust. The correlation matrix among the focal constructs for the two samples is
reported in Table 2. The result for structural path parameter estimates across the two samples is provided in Table 3.
Table 2
Correlation matrix of variables analyzed.
BTRUST
FTRUST
BCOM
BKNOW
BCOMPS
.575**
1
.513**
.424**
.520**
.602**
.509**
1
.574**
.748**
.719**
.457**
.700**
1
.568**
.677**
.590**
.643**
.656**
1
Automotive industry
Mean
3.087
Std.
0.979
2.700
0.846
2.199
0.858
2.761
0.906
2.503
0.917
Food industry
Mean
4.147
Std.
0.811
3.950
0.709
3.339
0.973
3.569
0.879
3.439
0.705
BTRUST
FTRUST
BCOM
BKNOW
BCOMPS
1
.494**
.519**
.681**
.665**
N = 345 for the automotive industry and N = 143 for the food industry.
The correlation matrix for the automotive industry sample is provided in the lower
diagonal, and the correlation matrix for the food industry sample is provided in the
upper diagonal.
**
Correlation is significant at the 0.05 level (2-tailed).
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
325
Table 3
Standardized parameter estimates of the tests of hypotheses.
Independent variables
Dependent variables
Trust of purchasing agent
Automotive industry
Strategic communication
Professional knowledge
Ability to compromise
Trust of purchasing agent
0.424**
0.458**
Relationship length
Supplier sales dependency
Trust of buying firm
Food industry
Automotive industry
Food industry
0.338**
0.211*
0.392**
0.536**
0.02
0.046
0.067
0.095
0.556**
0.285**
Overall fit index: Chi-square = 14.71, d.f. = 11, CFI = .995, GFI = .985; AGFI = .962, RMSEA = .034 for the automotive industry sample.
Overall fit index: Chi-square = 7.496, d.f. = 11, CFI = 1.00, GFI = .975, AGFI = .937, RMSEA = .000 for the food industry sample.
*
Parameters significant at 0.1 significance level.
**
Parameters significant at 0.05 significance level.
5.2.1. Testing of reciprocal relations between trust in the buying
firm and trust in the purchasing agent
We used structural equation modeling (SEM) to test the reciprocal relations between trust in the buying firm and trust in the
purchasing agent. SEM has become a commonly used technique
in management research to examine reciprocal relations between
two constructs using cross-sectional data (Wong and Law, 1999;
Mesquita et al., 2008). Wong and Law (1999) found that this
methodology results in parameter estimates that are a good proxy
for the true time-lagged effects with a reliability of p < 0.05. We
applied the method outlined in Wong and Law (1999) and Mesquita
et al. (2008) to test the reciprocal relationships between trust in
the buying firm and trust in the purchasing agent. Specifically, we
included different instrumental variables for each endogenous variable (e.g. trust in the buying firm and trust in the purchasing agent),
and correlated disturbance terms of each endogenous variable as
recommended by Wong and Law (1999).
We found that trust in the purchasing agent significantly
increases trust in the buying firm (b = 0.392, p < 0.05 for the automotive industry sample; and b = 0.536, p < 0.05 for the food industry
sample), which confirms H4a. Trust in the buying firm, however, did not have a significant effect on trust in the purchasing
agent (b = −0.055, p > 0.05 for the automotive industry sample; and
b = −0.720, p > 0.05 for the food industry). H4b was thus not supported. We therefore concluded that the relationship between trust
in the purchasing agent and trust in the buying firm is unidirectional from trust in the purchasing agent to trust in the buying firm.
We thus released the path from trust in the buying firm to trust in
the purchasing agent for further structural model testing.
5.2.2. Relationships between boundary spanning capabilities and
trust in the buying firm and in purchasing agents
The results of structural model testing are shown in Table 3. We
found that the structural model linking boundary spanning capabilities to trust has a satisfactory overall fit level (2 (11) = 14.71,
CFI = 0.995, GFI = 0.985; AGFI = 0.962, RMSEA = 0.034 for the automotive industry sample; 2 (11) = 7.496, CFI = 1.00, GFI = 0.975,
AGFI = 0.937, RMSEA = 0.000 for the food industry sample).
The relationship between a purchasing agent strategic communication and a supplier’s trust in the buying firm is positive
and significant (b = 0.338, p < 0.05 for the automotive industry sample; b = 0.211, p < 0.1 for the food industry sample), which supports
H1. Purchasing agent professional knowledge has a significant
and positive effect on a supplier’s trust of the purchasing agent
(b = 0.424, p < 0.05 for the automotive industry sample; b = 0.556,
p < 0.05 for the food industry sample). H2 is thus confirmed. As
hypothesized, purchasing agent ability to compromise also significantly increases a supplier’s trust of a purchasing agent (b = 0.458,
p < 0.05 for the automotive industry sample; b = 0.284, p < 0.05 for
the food industry sample), which supports H3. For the control
variables, no significant relationships were found between relationship length and a supplier’s trust of a buying firm (b = 0.020,
p > 0.05 for the automotive industry sample; b = 0.067, p > 0.05 for
the food industry sample) or between sales dependency and a
supplier’s trust of a buying firm (b = 0.046, p > 0.05 for the automotive industry sample; b = 0.095, p > 0.05 for the food industry
sample).
5.2.3. Testing of direct vs. mediating effects
We further compared our hypothesized model with a rival
model where direct paths are specified between all three types of
boundary spanning capabilities and the two types of trust (Bagozzi
and Yi, 1988; Wulf et al., 2001). As the two industry samples
share similar measurement model and path model structures, we
pooled the data to conduct the test of direct vs. mediating effects
(Bollen, 1989). We found that purchasing agent strategic communication significantly affects trust in the buying firm (b = 0.366,
p < .05), but no direct relationship was found between a purchasing agent’s strategic communication and trust in the purchasing
agent (b = −0.097; p > 0.05). Furthermore, professional knowledge
of a purchasing agent significantly increases trust in the purchasing
agent (b = 0.463, p < 0.05), but does not directly influence trust in the
buying firm (b = −0.022, p > 0.05). Similarly, a purchasing agent’s
ability to compromise has a significant and direct effect on trust
in the purchasing agent (b = 0.528, p < 0.05), but not on trust in
the buying firm (b = 0.041, p > 0.05). Adding the additional direct
effects did not contribute to a significant improvement in model
fit (2 = 1.412). We thus conclude that no evidence was found
for direct relationships between all three types of boundary spanning capabilities and both trust in the buying firm and trust in the
purchasing agent.
5.2.4. Robustness check
We conducted an additional test to examine whether the buying
companies have significant effects on the path coefficient estimates. To control for the buying company effect, we included a
dummy variable identifying the two buying companies. We found
that the dummy variable has significant effects on both trust in the
buying firm (b = .251, p < 0.05) and trust in the purchasing agent
(b = .100, p < 0.05). However, controlling for the buying company
effect did not change the significance levels and directions of the
path coefficient estimates.
Furthermore, we examine whether including multiple
responses from the same source creates inter-case dependence
that biases the study findings. We re-estimated the hypothesized
model including only a single response from each respondent. We
found no significant differences between the original data and this
subset of data in the directions and the significance levels of the
326
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
path coefficient estimates. The results of these tests increase our
confidence in the validity of the study findings.
6. Discussion and managerial implications
Building on boundary theory and supply chain cooperation
literature, we investigate how different types of boundary spanning capabilities of individual purchasing agents affect trust
development at multiple levels between buying and supplying
organizations. The findings show that a purchasing agent’s strategic
communication directly increases a supplier’s trust in the buying
firm, while a purchasing agent’s professional knowledge and ability
to compromise significantly influence a supplier’s trust in purchasing agents. The current research enriches the existing supply chain
trust literature (e.g., Doney and Cannon, 1997; Zaheer et al., 1998;
Perrone et al., 2003; Ireland and Webb, 2007; Hill et al., 2009)
by conceptualizing and empirically demonstrating the differential
effects of boundary spanning capabilities of purchasing agents on
buyer–supplier trust development at multiple levels.
We propose that strategic communication of purchasing agents
will significantly increase a supplier’s trust in the buying firm (H1).
The results show that strategic communication has a positive and
significant effect on a supplier’s trust in a buying firm, and the
relationships are consistent across both the automotive and food
industry samples. These findings suggest that strategic communication carried out by purchasing agents indeed can serve as a
vehicle for trust-building at the firm level. As proposed by Ireland
and Webb (2007), strategic communication of boundary spanners
provides a supplier intimate knowledge of a buying firm’s future
plans, which helps to coordinate the supply chain and increases a
supplier’s trust in the buying firm. The findings of this study support
their proposition by showing the direct effect of strategic communication on trust in a buying firm. Further, we found no significant
relationship between a purchasing agent’s strategic communication and a supplier’s trust in the purchasing agent. As the content
of strategic communication is largely delineated by the buying firm,
a supplier can reasonably attribute the capability of a purchasing
agent in distributing strategic communications to corporate policy
or training performed by the buying firm. In performing strategic
communication, purchasing agents primarily serve as a role-taker
by representing strategic directions and maintaining a trustworthy image for the buying firm (Perrone et al., 2003). As such, this
boundary spanning capability is closely linked with a buying firm’s
willingness to openly share information, which directly increases
trust in the buying firm. On the other hand, ineffective strategic
communication by purchasing agents can directly jeopardize a supplier’s trust in the buying firm as a whole.
Besides this direct mechanism of developing trust in the buying firm, we found that purchasing agents’ professional knowledge
enhances a suppliers’ trust in the purchasing agents (H2). These
results hold for both the automotive and food industry samples.
Furthermore, no direct effects were found between a purchasing
agent’s professional knowledge and a supplier’s trust in the buying firm. Professional knowledge of a purchasing agent signals his
or her overall competence and professional status within the firm.
Considered as a source of expert power held by individual boundary
spanners, professional knowledge is primarily linked with individual purchasing agents as opposed to a buying firm in the minds of
suppliers. As such, a supplier’s trust is primarily placed in the purchasing agent rather than the buying firm when a purchasing agent
demonstrates professional knowledge. Previous studies found that
general expertise (Doney and Cannon, 1997) and tenure within
an organization (Perrone et al., 2003) increase trust in the boundary spanner. The findings of this study extend those of Doney and
Cannon (1997) and Perrone et al. (2003) by showing that profes-
sional knowledge, including technical and commercial knowledge
and knowledge related to the supplier’s products and capabilities,
can significantly enhance a supplier’s trust in a boundary spanner.
Similar to professional knowledge, a purchasing agent’s ability
to reach compromise also directly increases a supplier’s trust in
the purchasing agent, but not trust in the buying firm. As Friedman
and Podolny (1992) propose, a boundary spanner is likely to risk
either being distrusted by supplier firms or by their own firms
if he or she fails to effectively negotiate a fair solution for issues
concerning the interests of both parties. Our findings support the
proposition of Friedman and Podolny (1992) by showing that a purchasing agent’s ability to reach compromise significantly increases
a supplier’s trust in the purchasing agent. This finding further
corroborates the observations of MacDuffie and Helper (2006) in
their case study of supplier relations in the automotive industry.
MacDuffie and Helper (2006) propose that in order for the automotive assemblers to build “collaboration with trust”, they need
to employ purchasing agents who adopt a team-orientation when
working with suppliers. A purchasing agent’s ability to compromise
builds the team-orientation required in collaborating with suppliers. Our findings extend those of MacDuffie and Helper (2006) by
showing that a purchasing agent’s ability to compromise directly
increases a supplier’s trust in the purchasing agent, which is then
transferred to inter-organizational trust between buying and selling organizations.
Furthermore, we found that a supplier’s trust in a purchasing
agent leads to a supplier’s trust in the buying firm but not vice versa.
Trust in individual boundary spanners is indeed an antecedent
of trust in the organization they represent. This finding suggests
that interpersonal relations can be sources of inter-organizational
relational rents. As trust is developed in an individual boundary
spanner before it is transferred to the firm, it is critical to maintain
the continuity of the interpersonal relationship in order to sustain
a long-term partnership with a particular firm. Surprisingly, we
did not find a significant relationship between trust in the buying firm and trust in individual purchasing agents. These findings
are different from the reciprocal relations on interpersonal trust
and inter-firm trust found by Doney and Cannon (1997). One possible explanation is that the samples used in our study concerns
existing relations between large buying firms and raw material or
component suppliers. These buying organizations have a number of
divisions and employ a large number of purchasing agents. Due to
the scale of the buying firms and variations among different divisions, a supplier may not consider trust in the buying firm as a
credible basis on which to infer trust in an individual purchasing
agent. Instead, they may prefer to use information acquired from
direct, personal experience with a purchasing agent as the basis
for their judgment of the trustworthiness of the purchasing agent.
Additionally, trust in the organization an individual belongs to is
used only when little direct experience with the individual exists
(Doney and Cannon, 1997). As one acquires direct experience with
a boundary spanning individual, he or she can make a judgment
based on these direct interactions. Because purchasing agents may
interact frequently with suppliers, trust in the buying firm may not
influence trust in purchasing agents in these relationships.
Managerially, the results show that suppliers develop trust in a
specific purchasing agent in addition to the buying firm. Because of
this, maintaining a relationship with a particular supplier imposes
restrictions on buying firms that assign purchasing agents to suppliers. If trust is developed in the individual, then maintaining
continuity of personal relationships takes on added importance.
Though firm-level relationships are categorized as ranging from
one-time transactions to long-term partnerships, a long-term partnership with a particular firm may not be sufficient to develop trust
if turnover of the purchasing agent assigned to the partnership is
high. The implication for managers is that they should be alert to
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
the importance of the personal relationships that develop between
boundary spanners of their own firms and those of the suppliers.
Maintaining trust may restrict the flexibility of managers in reassigning agents. Should reassignments need to be made, managers
should employ measures to reduce any loss of trust. These measures
may include finding a qualified replacement with professional
knowledge of the product market and technology. Additionally, the
outgoing purchasing agent trusted by suppliers should introduce
the new agent so that whatever trust that has been developed may
be transferred. Further, the outgoing purchasing agent can attempt
to transfer some knowledge and skills used to compromise with
suppliers to the new purchasing agent. In general, management
should ensure that the new purchasing agents have the boundary
spanning capabilities necessary to maintain trust with the supplier.
A second contribution for managers made by the study stems
from the capabilities themselves. The importance of strategic communication implies that purchasing agents need to be kept abreast
of the strategic plans of their organizations and be authorized and
willing to share this information with suppliers. The effect of professional knowledge suggests that agents may benefit from having
technical backgrounds or direct experience working for or with
suppliers. Such experience may allow these agents to better relate
to suppliers. Lastly, efforts to improve the negotiating and problemsolving skills of purchasing agents that focus on finding solutions
that are beneficial to both buying and supplying firms could be
justified.
7. Limitations and directions for future research
The findings of this study need to be interpreted in light of the
following limitations.
First, the current study restricts its focus to boundary spanners
who serve as purchasing agents. While purchasing agents can use
their positions to influence other departments and external entities and ensure that certain communications flow through them
(Strauss, 1962), they need not be the only contacts a supplier firm
has with the buying firm (Ireland and Webb, 2007). The value of this
research is thus in examining the role of purchasing agents on trustbuilding rather than a cataloging of all the boundary spanners who
may interact with a supplier. Furthermore, when several boundary
spanners from a buying firm such as engineering and purchasing
personnel interact with a supplier in supplier development programs, trust in the buying firm may be developed in different ways.
It is possible that the supplier judges the trustworthiness of each
of the firm’s personnel and forms a general opinion. Alternatively,
it may weigh the trustworthiness of particular boundary spanners
like purchasing agents who assume commercial responsibilities as
most important in estimating the credibility of a buying firm. These
research questions would be of interest to future development in
this area.
Additionally, when a buying firm engages in integration with
a supplier, this may work to mitigate the influence that a single purchasing agent exerts on trust-building in the buying firm.
Integration in general and strategic integration in particular (e.g.
Swink et al., 2007; Narasimhan et al., 2010) may give a supplier
an additional basis upon which to form its judgments. Thus, a valuable research question could be to what extent supplier integration
moderates the influence of boundary spanners on trust-building in
the buying firm.
Another limitation of the study concerns the study setting. We
collected data from two samples of suppliers of large buying organizations in the automotive and food industries. While using samples
from a single-industry controls for industry effects, supply chain
scholars (e.g. Noordewier et al., 1990; Jap, 1999) have cautioned
that the effects found in single-industry studies may be different
327
from other industry contexts. To address this concern, we collected data from firms in two distinct industries to cross-validate
the findings of the study. Although the use of two separate industry samples enables us to increase external validity of our study
findings and controls for potential industry effects, caution should
be used in generalizing the findings of the study to other industries with different power structures between buyers and suppliers.
For industries with a different power structure between buying
and supplying organizations, we consider it worthwhile for future
researchers to identify key variables and verify the research model
tested here in the specific industry context.
Third, data were collected from the supplier’s perspective in the
buyer–supplier relationship. We considered it appropriate for the
supplier salespeople a purchasing agent interacts with to evaluate
the boundary spanning capabilities of the purchasing agents. Furthermore, it is necessary to have the supplier to evaluate the trust
they place in the purchasing agents and the buying firm. The design
of this study thus makes it difficult to collect data from multiple
sources to minimize common method bias. This study, as with other
studies of this type (e.g. Takeishi, 2001; Jap and Anderson, 2003;
Griffith et al., 2006), would benefit from matched data between
different personnel within a supplier firm or personnel from both
suppliers and buying firms. Future research may explore other
methods to corroborate data collected from a single source to minimize or avoid concerns of common method bias. Additionally, we
collected the data at a single point in time. It may be valuable for
future research to collect the measures for the boundary spanning
capabilities and trust during different periods of time to establish
the causal sequence of the hypothesized relationships and further
minimize the potential influence of common method bias.
In conclusion, the focus of this study was to examine whether
boundary spanner capabilities enhance a supplier’s trust of an individual boundary spanner and the buying organization he or she
represents. The findings of our study demonstrate the importance
of retaining capable boundary spanners to build trust in industrial
supply chains. As such, we hope this works serves as a starting point
for future investigations of the role of individual boundary spanners
at generating relational rents for participating firms.
References
Adams, J.S., 1976. The structure and dynamics of behavior in organizational boundary roles. In: Dunnette, M.D. (Ed.), Handbook of Industrial and Organizational
Psychology. Rand McNally, Chicago, IL, pp. 1175–1199.
Aldrich, H., Herker, D., 1977. Boundary spanning roles and organization structure.
Academy of Management Review 2 (2), 217–230.
Anderson, J.C., Narus, J.A., 1990. A model of distributor firm and manufacturer firm
working partnerships. Journal of Marketing 54 (1), 42–58.
Armstrong, J.S., Overton, T.S., 1977. Estimating non-response bias in mail surveys.
Journal of Marketing Research 14 (3), 396–402.
Bagozzi, R.P., Yi, Y., 1988. On the evaluation of structural equation models. Journal
of the Academy of Marketing Science 16 (1), 74–94.
Bagozzi, R.P., Yi, Y., Phillips, L.W., 1991. Assessing construct validity in organizational
research. Administrative Science Quarterly 36 (3), 421–458.
Biong, H., Selnes, F., 1996. The strategic role of the salesperson in established
buyer–seller relationships. Journal of Business-to-Business Marketing 3 (3),
39–78.
Bollen, K.A., 1989. Structural Equations with Latent Variables. John Wiley & Sons,
New York.
Bradach, J.L., Eccles, R.G., 1989. Price, authority, and trust: from ideal types to plural
forms. Annual Review of Sociology 15, 97–118.
Cavusgil, S.T., Deligonul, S.Z., Zhang, C., 2004. Curbing foreign distributor opportunism: an examination of trust, contracts, and the legal environment in
international channel relationships. Journal of International Marketing 12 (2),
7–27.
Doney, P.M., Cannon, J.P., 1997. An examination of the nature of trust in buyer–seller
relationships. Journal of Marketing 61 (2), 35–51.
Dyer, J.H., Hatch, N.W., 2006. Relation-specific capabilities and barriers to knowledge
transfers: creating advantage though network relationships. Strategic Management Journal 27, 701–719.
Friedman, R.A., Podolny, J., 1992. Differentiation of boundary spanning roles: labor
negotiations and implications for role conflict. Administrative Science Quarterly
37 (1), 28–48.
328
C. Zhang et al. / Journal of Operations Management 29 (2011) 318–328
Gattiker, T.F., Huang, X., Schwarz, J.L., 2007. Negotiation, email, and Internet reverse
auctions: how sourcing mechanisms deployed by buyers affect suppliers’ trust.
Journal of Operations Management 25, 184–202.
Griffith, D.A, Harvey, M.G., Lusch, R.F., 2006. Social exchange in inter-organizational
relationships, the resulting benefits of procedural and distributive justice. Journal of Operations Management 24 (2), 85–98.
Handfield, R.B., Bechtel, C., 2002. The role of trust and relationship structure in
improving supply chain responsiveness. Industrial Marketing Management 31
(4), 367–382.
Hill, J.A., Eckerd, S., Wilson, D., Greer, B., 2009. The effect of unethical behavior
on trust in a buyer–supplier relationship: the mediating role of psychological
contract violation. Journal of Operations Management 27, 193–281.
IBM Business Consulting Services, 2004. Supply Chain Management, New Competitive Realities in the Automotive Value Chain Industry Report 1-32.
Ireland, R.D., Webb, J.W., 2007. A multi-theoretic perspective on trust and power in
strategic supply chains. Journal of Operations Management 25, 482–497.
Jap, S., 1999. Pie-expansion efforts: collaboration processes in buyer–supplier relationships. Journal of Marketing Research 36 (40), 461–475.
Jap, S., Anderson, E., 2003. Safeguarding interorganizational performance and
continuity under ex post opportunism. Management Science 49 (December),
1684–1701.
Johnson, D.W., Johnson, S., 1972. The effects of attitude intensity, expectation of
goal facilitation, and actual goal facilitation on interpersonal attraction. Journal
of Experimental Social Psychology 8 (3), 197–206.
Johnston, D.A, McCutcheon, D.M., Stuart, F.I., Kerwood, H., 2004. Effects of supplier
trust on performance of cooperative supplier relationships. Journal of Operations Management 22 (1), 23–38.
Kiessling, T., Harvey, M., Garrison, G., 2004. The importance of boundary-spanners
in global supply chains and logistics management in the 21st century. Journal of
Global Marketing 17 (4), 93–115.
Kraljic, P., 1983. Purchasing must become supply management. Harvard Business
Review 61 (September–October), 109–117.
Kumar, N.L., Stern, L.W., Anderson, J.C., 1993. Conducting interorganizational
research using key informants. Academy of Management Journal 36 (December),
1633–1651.
Laseter, T.M., Ramdas, K., 2002. Product types and supplier roles in product development, an exploratory analysis. IEEE Transactions on Engineering Management
49 (May), 107–118.
MacDuffie, J.P., Helper, S., 2006. Collaboration in supply chains with and
without trust. In: Adler, P. (Ed.), The Firm as Collaborative Community,
Re-constructing Trust in the Knowledge Economy. Oxford University Press,
Oxford.
Mesquita, L.F., Anand, J., Brush, T.H., 2008. Comparing the resource-based and relational views: knowledge transfer and spillover in vertical alliances. Strategic
Management Journal 29, 913–941.
McCutcheon, D., Stuart, F.I., 2000. Issues in the choice of supplier alliance partners.
Journal of Operations Management 18, 279–301.
Narasimhan, R., Swink, M., Viswanathan, S., 2010. On decisions for integration implementation: an examination of complementarities between product-process
technology integration and supply chain integration. Decisions Sciences 41 (2),
355–372.
Noordewier, T.G., John, G., Nevin, J.R., 1990. Performance outcomes of purchasing
arrangements in industrial buyer–vendor relationships. Journal of Marketing 54
(October), 80–93.
Paulraj, A., Lado, A.A., Chen, I.J., 2008. Inter-organizational communication as a
relational competency, antecedents and performance outcomes in collaborative buyer–supplier relationships. Journal of Operations Management 26 (1),
45–64.
Perrone, V., Zaheer, A., McEvily, B., 2003. Free to be trusted? Organizational
constraints on trust in boundary spanners. Organization Science 14 (4),
422–439.
Ring, P.S., Van De Ven, A.H., 1992. Structuring cooperative relationships between
organizations. Strategic Management Journal 13 (7), 483–498.
Sako, M., Helper, S., 1998. Determinants of trust in supplier relations: evidence from
the automotive industry in Japan and the United States. Journal of Economic
Behavior and Organization 34 (3), 387–417.
Schroeder, R.G., Linderman, K.L., Liedtke, C., Choo, A.S., 2008. Six sigma: definition
and underlying theory. Journal of Operations Management 26 (4), 536–554.
Smith, A.D., Plowman, D.A., Duchon, D., Quinn, A.M., 2009. A qualitative study of
high-reputation plant managers: political skill and successful outcomes. Journal
of Operations Management 27 (6), 428–443.
Spekman, R.E., 1979. Influence and information: an exploratory investigation of the
boundary role person’s basis of power. Academy of Management Journal 22,
104–117.
Stanley, L.L., Wisner, J.D., 2001. Service quality along the supply chain: implications
for purchasing. Journal of Operations Management 19, 287–306.
Stock, R.M., 2006. Interorganizational teams as boundary spanners between supplier
and customer companies. Journal of the Academy of Marketing Sciences 34 (4),
588–599.
Strauss, G., 1962. Tactics of lateral relationship: the purchasing agent. Administrative
Science Quarterly 7 (2), 161–186.
Swink, M., Narasimhan, R., Wang, C., 2007. Managing beyond the factory walls:
effects of four types of strategic integration on manufacturing plant performance. Journal of Operations Management 25 (1), 148–164.
Takeishi, A., 2001. Bridging inter- and intra-firm boundaries: management of supplier involvement in automobile product development. Strategic Management
Journal 22 (May), 403–433.
Tushman, M.L., Scanlan, T.J., 1981. Characteristics and external orientations of
boundary spanning individuals. Academy of Management Journal 24 (1),
83–98.
Wong, C.-S., Law, K.S., 1999. Testing reciprocal relations by nonrecursive structural
equation models using cross-sectional data. Organizational Research Methods
2 (1), 69–87.
Wulf, K.D., Odekerken-Schroder, G., Iacobucci, D., 2001. Investments in consumer
relationships: a cross-country and cross-industry exploration. Journal of Marketing 65 (October), 33–50.
Zaheer, A., McEvily, B., Perrone, V., 1998. Does trust matter? Exploring the effects
of interorganizational and interpersonal trust on performance. Organization
Science 9 (2), 141–159.
Zhang, C., Henke Jr., J.W., Griffith, D.A., 2009. Do buyer cooperative actions matter
under relational stress? Evidence from Japanese and U.S. assemblers in the U.S.
automotive industry. Journal of Operations Management 27, 479–494.